Comparative Performance of Different Adaptive Cdma Forms on Dispersive Channels

Spread spectrum techniques have been used for many cellular communication systems, and are being adopted widely for military communication networks. The motivation for employing this technique is its ability to suppress fading due to multipath propagation. In this paper, receivers based on chip level equalization are considered for a downlink code division multiple access (CDMA) system employing user-specific long or short spreading sequences to ensure good performance at low signal-to-noise ratios (SNRs). Adaptive turbo architectures for joint parameter estimation and multiple-access interference cancellation integrate the functions of equalization, multi-access interference cancellation, error correction and phase-carrier tracking. A direct adaptive CDMA receiver uses continuous pilots as training data and the filter coefficients are optimized based on the minimum mean square error (MMSE) criterion to suppress noise, ISI, and multiple access interference (MAI) with reasonable system complexity. Moreover, the proposed detectors with continuous pilots operate just in a tracking mode after the initial delay acquisition is accomplished. By exploiting the presence of a continuous pilot signal in the entire transmitted frame, this approach is well suited to tracking fast fading channels. The receivers are compared with Rake multiuser receiver-based channel estimation (CE) techniques using data obtained from experimental trials in the North Sea. Furthermore, three simultaneous users are reliably acquired and demodulated using a few stages. These results show that the direct adaptive receivers in both types of CDMA exhibit better immunity against the spectral channel characteristics and significant performance improvement over CE based receivers.